Mutations of Epstein-Barr Virus gH That Are Differ
http://www.100md.com
病菌学杂志 2005年第17期
School of Biological Sciences, University of Missouri—Kansas City, Kansas City, Missouri
Department of Microbiology and Immunology, Center for Molecular and Tumor Virology, and Feist-Weiller Cancer Center, Louisiana State University Health Sciences Center, Shreveport, Louisiana
ABSTRACT
The core fusion machinery of all herpesviruses consists of three conserved glycoproteins, gB and gHgL, suggesting a common mechanism for virus cell fusion, but fusion of Epstein-Barr virus (EBV) with B cells and epithelial cells is initiated differently. Fusion with B cells requires a fourth protein, gp42, which complexes with gHgL and interacts with HLA class II, the B-cell coreceptor. Fusion with an epithelial cell does not require gp42 but requires interaction of gHgL with a novel epithelial cell coreceptor. Epithelial cell fusion can be inhibited by gp42 binding to gHgL and by antibodies to gH that fail to block B-cell fusion. This suggests that regions of gHgL initiating fusion with each cell are separable from each other and from regions involved in fusion itself. To address this possibility we mapped the region of gH recognized by a monoclonal antibody to gH that blocks EBV fusion with epithelial cells but not B cells by making a series of chimeras with the gH homolog of rhesus lymphocryptovirus. Proteins with mutations engineered within this region included those that preferentially mediate fusion with B cells, those that preferentially mediate fusion with epithelial cells, and those that mediate fusion with neither cell type. These results support the hypothesis that the core fusion function of gH is the same for B cells and epithelial cells and that it differs only in the way in which it is triggered into a functionally active state.
INTRODUCTION
Epstein-Barr virus (EBV) is a human herpesvirus that predominantly infects B lymphocytes and epithelial cells. It causes infectious mononucleosis, an immunopathology initiated by an expansion of virus-infected B cells, immunoblastic lymphomas of the immunosuppressed, which are driven by virus infection, and oral hairy leukoplakia, an epithelial lesion produced as a result of lytic replication. The virus is also associated with additional lymphoid and epithelial malignancies, including Burkitt's lymphoma, Hodgkin's disease, and nasopharyngeal and gastric carcinomas (22). In normal healthy individuals the B cell is the primary reservoir of infection, but there is mounting evidence that epithelial cells are at least periodically infected during long-term carriage (21). Shuttling of virus between the two cell types may play a role in efficient initiation of infection of a new host and in replenishment of the infected B-cell pool (3, 26, 27).
Infection of B cells and epithelial cells occurs by different routes (16), and the subset of envelope proteins involved in mediating virus fusion with the two cell types is different. Following attachment via gp350 and CR2, fusion with a B cell requires glycoprotein gB and an oligomeric three-part complex of glycoproteins, gH, its chaperone gL, and gp42 (9). gB and gHgL are essential to the fusion machinery of all known herpesviruses (24); gp42 and its homologs are unique to EBV and other 1 herpesviruses (5, 13, 23). gp42 interacts with HLA class II on the B-cell surface (25), and this interaction is essential for triggering the fusion machinery into an active state (9, 10, 12, 13, 17, 28).
The requirements for fusion with an epithelial cell are less well understood but are in some respects fundamentally different. The core fusion machinery comprised of gHgL and gB is probably the same (15), but an interaction between gp42 and HLA class II is not used for its activation. Instead, gH, as part of a two-part gHgL oligomer, interacts directly with an as-yet-unknown molecule(s) in the epithelial membrane. This interaction, and fusion of virus with an epithelial cell, is impeded by the presence of gp42 or by a number of monoclonal antibodies (MAbs) to gH or gHgL that have no effect on infection of B cells (4, 18, 29). Thus, while only three-part complexes that include gp42 can participate in fusion with B cells, only two-part complexes that lack gp42 can participate in epithelial cell fusion.
EBV virions carry both three-part and two-part complexes, and modulation of the amounts of gp42 in the virion, which alters the ratio between the two, influences the tropism of the virus. Virions made in HLA class II-positive B cells lose some three-part complexes as a result of an intracellular interaction with HLA class II that targets the complexes to a degradative pathway. Virions made in HLA class II-negative epithelial cells lose none and carry a higher amount of gp42. Epithelial-derived virions are considerably better at infecting B cells, whereas B-cell-derived virions are somewhat better at infecting epithelial cells (3, 4).
If the core fusion machinery of EBV for both cell types were comprised of gB and gHgL and differed only in the way in which gHgL is triggered into a functionally active form, then it should be possible to derive a gHgL complex that is capable of supporting fusion with an epithelial cell but not a B cell, or vice versa. To test this hypothesis we first mapped the region of gH that is recognized by one of the MAbs that blocks epithelial but not B-cell infection by taking advantage of its failure to react with the gH homolog of the closely related rhesus lymphocryptovirus (Rh-LCV). A series of insertion mutations was made in this region in EBV gH, and each was tested for expression and the ability to mediate fusion with a B-cell or an epithelial cell membrane. We report here on the phenotypes of these mutated proteins, which include those that, although expressed at the cell surface, have either completely or only differentially lost the ability to mediate fusion with B cells and epithelial cells.
MATERIALS AND METHODS
Cells. CV-1 monkey kidney cells were grown in Dulbecco's modified Eagle's medium. AGS, a human gastric carcinoma cell line, and Chinese hamster ovary cells (CHO-K1) were grown in Ham's F-12 medium. Daudi 29 cells (a gift of Richard Longnecker, Northwestern University), B cells that stably express T7 RNA polymerase from the pOS2 vector (30), were grown in RPMI medium. All culture media were obtained from Gibco-BRL Life Technologies (Grand Island, NY) and supplemented with 10% heat-inactivated fetal bovine serum.
Antibodies. Antibodies used were MAbs E1D1 to gHgL, CL59 and CL40 to gH, CL55 to gB, F-2-1 to gp42 (13), and M2 (Sigma-Aldrich, St. Louis, MO) to the Flag epitope. All noncommercial antibodies were affinity purified on protein A-agarose.
Virus. Stocks of recombinant vaccinia virus expressing the T7 RNA polymerase (vvT7; a gift of William Britt, University of Alabama, Birmingham) were grown in CV-1 cells infected at a multiplicity of infection of 0.01. When all cells appeared swollen (approximately 48 h postinfection), they were shaken or scraped into the medium, pelleted by centrifugation, and resuspended in 2 ml of clarified medium per 30 million cells. The resuspended cells were subjected to three cycles of freeze-thawing, sonicated for 45 s in 4-ml aliquots in round-bottom tubes, clarified by centrifugation, pooled, and stored at –85°C.
Plasmid constructs for epitope mapping. Plasmids pTM1-EBV gH and pTM1-EBV gL (13) were made by cloning PCR-amplified sequences into the pTM1 vector (19), which contains a T7 promoter, the encephalomyocarditis virus cap-independent translation signal, a multiple cloning site, and a T7 transcriptional terminator. Translation initiates at the NcoI site in the multiple cloning region. In pTM1-EBV gH this site is lost as a result of blunt-end cloning. Plasmid pTM1-EBV119-Rh-LCV gH was made by cutting a 2.8-kb HindIII fragment encompassing the open reading frame encoding the gH homolog of Rh-LCV from the DK12 cosmid clone of Rh-LCV DNA (a gift of Fred Wang, Harvard University) (23). The fragment was blunt ended, recut with NcoI, and cloned into pTM1-EBV gH cut with NcoI and HincII. The first 119 bp of the insert represented EBV gH sequence, and the remaining 2,004 bp represented Rh-LCV gH sequence, including the stop codon and an additional 23 bp of 3' sequence. The EBV gH amino acid sequence encoded by the 119 bp of EBV DNA differed from Rh-LCV gH at positions 2, 7, 10, 13, 15, 17, 33, and 35. Only residues 33 and 35 are predicted to remain after cleavage of the signal sequence at residue 17. These residues were changed to those of Rh-LCV gH by amplification of a fragment extending from the NcoI site to a ClaI site upstream of the gH start in pTM1-EBV gH. The changes were incorporated in the primer encompassing the NcoI site to make pTM1-Rh-LCV gH. Further chimeras of EBV gH and Rh-LCV gH were constructed by insertion of PCR-amplified fragments of EBV DNA in place of equivalent fragments of Rh-LCV gH either by using existing shared restriction sites (NcoI at bp 119) or by creating a new site in the EBV gH DNA sequence, which did not alter coding sequence, to correspond with existing sites of Rh-LCV gH.
Transfection-infection protocol. CV-1 cells were grown to 90% confluence in 100-mm-diameter petri dishes and infected with vvT7 at a multiplicity of infection of 5. Thirty minutes later, the inoculum was removed, and the cells were washed twice in medium without serum and transfected with pTM1 plasmids. Ten micrograms of DNA was mixed with 30 μl of Lipofectamine (Gibco-BRL) made to a total volume of 200 μl with serum-free medium.
EBV plasmid constructs for fusion assays. EBV gH, EBV gL, EBV gB, and EBV gp42 were amplified by PCR and cloned into the pCAGGS/MCS vector (a gift of Martin Muggeridge, Louisiana State University Health Sciences Center) for expression under control of the -actin promoter in cooperation with the human cytomegalovirus immediate-early enhancer (20). These constructs were designated pCAGGS-gH, pCAGGS-gL, pCAGGS-gB, and pCAGGS-gp42, respectively. A Flag-tagged version of pCAGGS-gH, pCAGGS-Flag-gH, was made by insertion of an eight-amino-acid Flag epitope between residues 22 and 23 of the EBV gH sequence, four residues after the predicted cleavage site of the putative signal sequence. For construction of mutations in pCAGGS-Flag-gH, a 1.6-kb BglII-PvuII fragment encompassing sequence coding for residues 500 to 629 was cloned into the 2.5-kb pSP72 vector (Promega, Madison, WI) to make pSP72-EBV gH1.6. This construct was used as the template for a series of insertion mutants made using the QuikChange II protocol (Stratagene, La Jolla, CA). Primers for insertion mutagenesis were designed to introduce in-frame insertions of the sequence TCG CGA, which encodes serine and arginine residues and introduces a novel and unique NruI restriction site. DNA was amplified for 18 cycles (60 s at 95°C, 90 s at 60°C, and 9 min at 72°C), and the template was eliminated by digestion with DpnI. Insertions were initially screened by digestion with NruI. XhoI-NsiI fragments encompassing the mutated BglII-PvuII fragments were recloned back into pCAGGS-Flag-gH for assay of fusion activity or pTM1-EBV gH for biochemical analysis. The desired insertion or mutation was confirmed by sequencing of the fragments returned to pCAGGS-Flag-gH.
Immunofluorescence. For internal staining, slides bearing air-dried acetone-fixed cells were incubated at 37°C for 30 min in a humidified atmosphere with MAbs, washed three times with phosphate-buffered saline for 5 min each, reincubated for 30 min with the appropriate dilution of fluorescein isothiocyanate-conjugated sheep anti-mouse immunoglobulin (ICN/Cappel, Irvine, CA), washed three times, and mounted in mounting medium (Kirkegaard & Perry Laboratories, Gaithersburg, MD). For surface staining, unfixed cells were sequentially incubated with MAbs and fluorescein isothiocyanate-conjugated anti-mouse immunoglobulin. Cells were washed three times between incubations and three times before mounting in mounting medium.
Levels of cell surface expression. The levels of expression of wild-type and mutant gH constructs at the cell surface were measured by flow cytometric analysis of CHO-K1 cells transfected with pCAGGS-Flag-gH or gH mutants and pCAGGS-gL. A 3.2-μg aliquot of DNA was mixed with 4 μl Target transfectin F2 and 12 μl Target peptide enhancer (Targeting Systems, Santee, CA) in high-glucose Dulbecco's modified Eagle's medium. Twenty-four hours later, transfected cells were washed with ice-cold phosphate-buffered saline containing 2% fetal bovine serum and serially reacted with MAb M2 (anti-Flag) and phycoerythrin-conjugated anti-mouse immunoglobulin (Jackson ImmunoResearch Laboratories, West Grove, PA), with washing between each step.
AGS cell-cell fusion assay. AGS cells were seeded in two-well chamber slides and were transfected at 70 to 80% confluence for 4 h with 0.25 μg pCAGGS-Flag-gH or 0.25 μg mutated pCAGGS-Flag-gH together with 0.25 μg pCAGGS-gL and 0.6 μg pCAGGS-gB. Plasmids were mixed with 1 μl Target transfectin F2 and 3 μl Target peptide enhancer in high-glucose Dulbecco's modified Eagle's medium. Twenty-four hours posttransfection cells were fixed with ice-cold acetone and stained with MAb to gB. Cells expressing gB and containing four or more nuclei were considered to have undergone fusion and were recorded as one fusion event. The extent of fusion was calculated as the number of fusion events as a percentage of the total number cells of expressing gB. No fusion was seen in cells that did not stain with MAb to gB.
Epithelial cell-B-cell fusion assay. CHO-K1 cells were seeded in six-well plates and were transfected at 70 to 80% confluence for 4 h with 0.8 μg pCAGGS-Flag-gH or mutated pCAGGS-Flag-gH together with 0.8 μg gL, 1.4 μg pCAGGS-gB, and 1.2 μg pCAGGSgp42 as well as 0.8 μg pST7-Luc containing the T7 promoter upstream of the luciferase gene (6) (a gift of Martin Muggeridge). Each well of transfected cells was then overlaid with two million Daudi 29 cells that expressed T7 RNA polymerase. Twenty-four hours later, cells were washed twice with phosphate-buffered saline and lysed with 500 μl passive lysis buffer (Promega). Luciferase substrate (100 μl) was added to 20 μl supernatant of lysate. Luminosity readings were obtained by using a TD-20/20 luminometer (Promega).
Radiolabeling and immunoprecipitation. CV-1 cells infected with vvT7 and transfected with pTM1 plasmids were labeled biosynthetically with 100 μCi of Pro-Mix (70% [35S]methionine, 30% [35S]cysteine) and 300 μCi of [35S]cysteine (>1,000 Ci/mmol; Amersham Biosciences, Arlington, IL) per dish (approximately 107 cells) as previously described (13). Labeled cells were solubilized in radioimmunoprecipitation buffer (50 mM Tris-HCl, pH 7.2, 0.15 M NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate [SDS], 0.1 mM phenylmethylsulfonyl fluoride, and 100 U of aprotinin per ml) and immunoprecipitated with antibody and protein A-Sepharose CL4B (Sigma). Immu-noprecipitated proteins were washed, dissociated by boiling for 2 min in sample buffer with 2-mercaptoethanol, and analyzed by SDS-polyacrylamide gel electrophoresis in 10% polyacrylamide cross-linked with 0.28% N,N'-diallyltartardiamide followed by fluorography.
RESULTS
The region of gH that interacts with MAb CL59 lies between residues 500 and 629. MAbs E1D1, CL40, and CL59 all inhibit EBV infection of epithelial cells and, since none of the antibodies neutralize B-cell infection, the epitopes that they recognize are uniquely important to infection of epithelial cells (13, 18). MAb E1D1 reacts with EBV gH only if it is expressed together with EBV gL, and MAbs CL40 and CL59 react with gH alone. To begin to map the regions of gH involved in the generation of these epitopes, we explored the reactivity of the MAbs with the gH homolog of Rh-LCV, which has 86% similarity to EBV gH (23). EBV gL was expressed together with Rh-LCV gH in the pTM1 vector for expression under control of the T7 promoter. The plasmids were transfected into CV1 cells that were also infected with recombinant vaccinia virus expressing the T7 polymerase. Transfected and infected cells were acetone fixed and tested for reactivity with MAbs in indirect immunofluorescence assays. MAbs E1D1 and CL40 both reacted with Rh-LCV-gH, but MAb CL59 did not (Fig. 1).
In order to map the domain recognized by MAb CL59, a series of chimeric constructs were made in which different fragments of EBV gH replaced linearly equivalent fragments of Rh-LCV gH. The ability of each chimera to be recognized by the antibody was examined by indirect immunofluorescence. EBV residues 1 to 39, 40 to 319, or 320 to 457 did not confer reactivity with MAb CL59, but residues 458 to 628 did (Fig. 1), indicating that the MAb CL59 epitope resides within this fragment of EBV gH (Fig. 2). Reactivity was retained when only EBV gH residues 501 to 628, which have 74.7% identity with the equivalent Rh-LCV gH fragment (Fig. 3), were replaced. In an attempt to further narrow down the region recognized by CL59, two final chimeric constructs were made, one in which only residues 501 to 589 were replaced with EBV sequence and one in which only residues 590 to 628 were replaced. Neither of these two constructs was recognized by MAb CL59. It is formally possible that the epitope recognized by MAb CL59 is linear and spans the junction of the two fragments. However, since MAb fails to react with EBV gH on Western blots (data not shown), it is perhaps more likely that it recognizes a nonlinear epitope formed by folding of residues 501 to 628.
Expression of EBV gH with insertion mutations between residues 500 and 629. Mapping of the reactivity of MAb CL59 to residues 501 to 628 of EBV gH identified a region that was potentially of unique importance to epithelial infection. To test this possibility a series of 13 insertion mutations were made within the region (Fig. 3) and built into an EBV gH construct that had been cloned into the pCAGGS vector for expression under control of the -actin promoter (pCAGGS-Flag-gH). Because it was possible that some of the insertions might affect the ability of the protein to be recognized by all of the gH-specific MAbs, sequence encoding a linear Flag epitope was also inserted at a position that would put it two residues after the predicted cut site of the putative signal sequence (11). Each mutational insertion consisted of two amino acids, serine and arginine, since a similar strategy has proven viable for identifying functional regions of another conformationally sensitive herpesvirus glycoprotein, glycoprotein gB of herpes simplex virus type 2 (M. Muggeridge, personal communication).
The expression of each construct, together with pCAGGS-gL, was first examined by indirect immunofluorescence staining of acetone-fixed cells. Although all constructs were expressed as judged by reactivity with the anti-Flag antibody, they fell phenotypically into three groups (Table 1). Numbers 1 to 6 and 9 to 11 reacted well not only with anti-Flag but also with all three EBV-specific MAbs. Numbers 7 and 13 reacted with the same antibodies but more weakly. Numbers 8 and 12, which also reacted only weakly with anti-Flag antibody, failed to react at all with any of the EBV-specific MAbs. To determine which, if any, of the mutants could be expressed at the cell surface, indirect immunofluorescence staining was repeated with unfixed cells. The constructs again fell into three broad groups (Table 2). Group one mutants, including numbers 1 to 5 and 9 to 11, were readily detected at wild-type levels at the cell surface with both anti-Flag and all the EBV-specific MAbs. Group two mutants, including number 6, which had reacted well in fixed cells, and numbers 7 and 13, all reacted only weakly at the cell surface with anti-Flag and the EBV-specific MAbs. Number 6 contained an insertion that interrupted an N-linked glycosylation site and had a faster mobility when analyzed by electrophoresis and Western blotting (not shown). Group three, including numbers 8 and 12, was not seen by anti-Flag or any of the EBV-specific MAbs.
A more quantitative assessment of the relative levels of expression at the cell surface of each of the constructs made by flow cytometric analysis of unfixed cells stained with M2 anti-Flag antibody revealed some more subtle heterogeneities. Group one could now be divided into numbers 3, 5, 9, and 10, which were expressed at levels similar to wild-type pCAGGS-Flag-gH, and numbers 1, 2, 4, and 11, which were expressed less well but still at more than 50% of wild-type levels (Fig. 4a). Group two, numbers 6, 7, and 13, was expressed at approximately 25% of wild-type levels. As before, no cell surface expression was seen of insertion mutants numbers 8 and 12, both of which failed to react with any of the EBV-specific MAbs.
Comparison of the ability of wild-type gH and insertion mutants to mediate cell-to-cell fusion of epithelial cells. pCAGGS-Flag-gH and each of the insertion mutants were next examined for the ability to mediate cell-to-cell fusion when transfected into AGS epithelial cells together with pCAGGS-gL and pCAGGS-gB. As expected, no fusion was seen in cells transfected with pCAGGS-gB alone, cells transfected with pCAGGS-Flag-gH and pCAGGS-gL, or cells transfected with pCAGGS-gB and pCAGGS-gL. However, when pCAGGS-gB, pCAGGS-Flag-gH, and pCAGGS-gL were coexpressed, multinucleate cells were seen (Fig. 5). A count of the number of cells that were successfully transfected as indicated by reactivity with MAb CL55 to gB and the number of foci of fluorescence in which more than four nuclei were present indicated that pCAGGS-Flag-gH supported fusion of 49 ± 2% (mean ± standard deviation) of the total number of transfected cells. A pCAGGS-gH construct expressing wild-type gH without insertion of the Flag tag mediated a similar degree of fusion (data not shown).
The fusion mediated by each insertion mutant was expressed as a percentage of fusion supported by wild-type pCAGGS-Flag-gH. None induced fusion as efficiently as wild-type gH, but numbers 3, 5, and 10, all of which were expressed at wild-type levels, induced fusion to a level of at least 50% of wild-type pCAGGS-Flag-gH (Fig. 4b). Numbers 1 and 2, which were expressed at more than 50% of wild-type levels, and number 13, which was expressed at about 25% of wild-type levels, induced lower but still significant levels of fusion. Numbers 4, 6, 7, 8, 9, 11, and 12 induced little or no fusion despite the fact that all but two, 8 and 12, were recognized by MAbs E1D1, CL40, and CL59 and two, numbers 9 and 11, were expressed at wild-type or close-to-wild-type levels, respectively.
Comparison of the ability of wild-type gH and insertion mutants to mediate cell-cell fusion with target B cells. Fusion of B cells which grow in suspension is not readily detected by microscopy. As an indirect measure of the ability of wild-type gH and insertion mutants to mediate fusion with a B cell, we therefore used an assay previously developed for this purpose by Haan and colleagues (9). Constructs were transfected into CHO-K1 cells, which do not fuse after expression of EBV gHgL and gB (15), together with pCAGGS-gL, pCAGGS-gB, pCAGGS-gp42, and pST7-Luc. Transfected cells were subsequently overlaid with Daudi 29 B cells which expressed the T7 polymerase. Fusion was determined as the increase in luciferase activity relative to a control in which pCAGGS-Flag-gH or its mutated derivatives were replaced with empty vector. The degree of fusion mediated by each insertion mutant was then expressed as a percentage of that induced by wild-type pCAGGS-Flag-gH.
Three mutants, numbers 3, expressed at wild-type levels, number 11, expressed at approximately 75% of wild-type levels, and number 13, expressed at approximately 25% of wild-type levels, induced levels of fusion that were variable but in a series of 12 experiments proved to be not significantly different from wild type (Fig. 4c). Number 5, which was expressed at wild-type levels, supported fusion to a level of only 50% of wild type. Lower levels of fusion again were mediated by numbers 1 and 2, which were expressed at between 50 and 75% of wild-type levels. Numbers 6 to 10 and 12 supported little or no fusion. Of these, numbers 9 and 10 were expressed at wild-type levels, 6 and 7 were expressed at about 25% of wild type, and 8 and 12 were not detectable at the cell surface at all.
Interaction of insertion mutants with gp42. A comparison of the levels of cell surface expression and abilities to mediate fusion with epithelial cells or B cells highlighted three particularly interesting insertion mutants, numbers 9, 10, and 11, which, although expressed at wild-type or close-to-wild-type levels, had very different phenotypes. Number 10 preferentially supported fusion with epithelial cells, number 11 preferentially supported fusion with B cells, and number 9 supported fusion with neither. Wild-type gH, if associated with gL, forms a three-part complex of gH, gL, and gp42 and can be immunoprecipitated by MAb F-2-1 to gp42. To determine if the low levels of B-cell fusion mediated by numbers 9 and 10 reflected an inability to interact with gp42, the same mutations were built into pTM1-EBV gH for higher levels of expression. Cells transfected with wild-type or mutated pTM1-EBV gH together with pTM1 gL and pTM1 gp42 were radiolabeled, and proteins were immunoprecipitated with MAb F-2-1 to gp42. F-2-1 immunoprecipitated not only wild-type pTM1-EBV gH but also the mutated proteins, indicating that despite their compromised ability to support B-cell fusion they both retained the ability to reform the three-part complex of gH, gL, and gp42 (Fig. 6).
DISCUSSION
Entry of all herpesviruses into cells requires fusion of the virus envelope with a cell membrane, either at the cell surface or after endocytosis. The ways in which this is accomplished are not completely understood for any herpesvirus, but some common themes are emerging (reviewed in reference 24). Three glycoproteins, gB, gH, and gL, which have homologs in every herpesvirus so far studied, are essential and make up the core fusion machinery. Triggering this machinery into action requires interaction with a coreceptor or entry mediator. Thus, triggering fusion of herpes simplex virus with many cells requires an interaction between a fourth glycoprotein, glycoprotein gD, and one of several cell entry receptors. Triggering of fusion of human herpesvirus 8 probably requires an interaction between gB and integrins (1). EBV is, however, perhaps unusual in two ways. First, although fusion with a B cell requires a fourth protein, gp42, which interacts with the entry mediator or coreceptor HLA class II, fusion with an epithelial cell does not. Second, gp42, which is essential for B-cell fusion, interferes with epithelial cell fusion, and MAbs to gHgL that block epithelial cell fusion have no effect on fusion with a B cell (4, 29). The understanding that gHgL are essential for fusion with both B cells and epithelial cells, despite the fact that the activation of fusion is different, was our impetus to look for regions of the complex that are important to function as part of the core fusion machinery and regions that are separably involved in triggering B cell and epithelial cell fusion. Since we had made three MAbs that inhibited epithelial cell entry but not B-cell entry, mapping their reactivities seemed a rational approach to begin the search for such regions. We were fortunate to find not only that only two of the three MAbs reacted with the gH homolog of the very closely related Rh-LCV when it was expressed together with EBV gL, but also that the MAb that failed to react was one that recognizes EBV gH in the absence of gL. Its epitope could then conclusively be said to be carried by gH and chimeras of Rh-LCV gH, and EBV gH could be used locate its general position. The region of gH carrying the epitope, which we were unable to narrow down further than a 129-amino-acid stretch, was in the carboxy-terminal third of the protein, only 52 amino acids from the predicted transmembrane domain. It does not overlap with the equivalent domain of herpes simplex virus gH that has been identified as a potential fusion protein (8). However, it is a region of gH that has been implicated in fusion of both the alphaherpesvirus herpes simplex virus (7), and the betaherpesvirus human herpesvirus 6 (2, 14). The insertion mutations described here confirm its functional importance.
Three insertion mutants, numbers 9, 10, and 11, were of particular interest, because although each was expressed at wild-type or close-to-wild-types levels, each had a very different functional phenotype. Number 9 supported fusion with neither B cells nor epithelial cells and would appear to confirm that, as for other herpesviruses, this carboxy-terminal region of gH is critical for function of the core fusion machinery. Number 10 preferentially supported epithelial cell fusion, whereas number 11 supported wild-type levels of fusion with B cells and little or no fusion with epithelial cells. The failure of number 11 to support epithelial cell fusion could not simply be ascribed to its expression level, since other insertion mutants, such as numbers 1, 2, and 13, were expressed at similar or lower levels and supported significant fusion. This suggests that, as expected, the triggers provided by gp42 interacting with HLA class II and by gHgL interacting with the epithelial cell coreceptor are mediated by separate structural features of gHgL. Further analysis of mutant number 11 will be needed to determine if the insertion has interfered with the ability of virus to bind to epithelial cells via a gHgL coreceptor (4, 18). Since both insertion mutants numbers 9 and 10 retained their interaction with gp42, both are likely to have retained the ability to interact indirectly with the B-cell coreceptor HLA class II. The fairly close proximity of insertions 10 and 11 suggests that the structures required for a functional interaction with either a B cell or an epithelial cell coreceptor may be proximal. This would not be altogether surprising, since complexes of gHgL that are saturated with gp42 cannot support epithelial cell fusion (29) and a soluble form of gp42 that can bind in trans impedes the interaction between gHgL and a specific epithelial receptor, possibly the coreceptor needed for fusion (4). MAb CL59 fails to block B-cell fusion, and addition of soluble gp42 to virus does not reduce the ability of the virus to bind CL59, as judged by flow cytometry (unpublished data). However, this could be explained if the antibody had a lower affinity for gHgL than did gp42 and was thus unable to compete with gp42 for binding to the same region of the gHgL complex.
The majority of the insertion mutations that were made preserved the ability of the protein to be recognized by each of the MAbs specific for gHgL which recognize conformational epitopes. In addition, all of those recognized by these MAbs were expressed at least at some level at the cell surface. This suggests that each was able to interact with gL. The association was not formally evaluated, since it is well established for all herpesviruses, including EBV (31), that without such an interaction gH is not authentically processed and transported to the cell surface.
Not surprisingly, mutants that were not expressed at all at the cell surface were also unable to support fusion with epithelial cells, but beyond this there was no simple relationship between levels of expression and levels of fusion with epithelial cells. Thus, of the four mutants, numbers 3, 5, 9, and 10, which were expressed at wild-type levels, three supported fusion at greater than 50% of wild-type levels and one supported no fusion at all. None of the mutants, even those expressed at wild-type levels, were able to support epithelial cell fusion as extensively as the wild-type protein. It is impossible to know, without a structural model of gHgL, exactly what this reflects, but it suggests that perhaps even though the insertion mutants are authentically transported and are reactive with MAbs known to recognize conformational epitopes, they are at least to some degree either initially misfolded or unable, subsequently, to assume fully a conformation required for switching to a functionally active state.
In contrast, three of the insertion mutants, numbers 3, 11, and 13, supported fusion with B cells at levels that were not significantly different from wild type and another, number 5, supported fusion at a level greater than 50% of wild type. The assays used to estimate fusion with the two different cell types are clearly different. It is formally possible that the luciferase assay is in some cases measuring pore formation in the absence of the pore expansion that would be necessary for production of multinucleate cells that were scored as epithelial fusion. However, the luciferase assay is not simply overestimating the ability of a given protein to support fusion since, as is clear from the composite analysis, there were two insertion mutants, numbers 5 and 10, which scored higher in the epithelial assay than the B-cell assay.
In the absence of a crystal structure for gHgL, it is not possible to understand the full impact of the mutations that have been made. However, three points seem clear. First, that the carboxy-terminal region of gH is critical to the support of fusion with either B cells or epithelial cells and is thus relevant to the function of the core fusion machinery. Second, that the same region is relevant to the trigger by which gHgL becomes functional. Third, that although mutations in this region can interfere with triggering via HLA class II or the epithelial cell coreceptor, the sequences important for the triggering events are distinct.
ACKNOWLEDGMENTS
This work was supported by Public Health Services grants AI20662 from the National Institute of Allergy and Infectious Diseases and DE016669 from the National Institute of Dental and Craniofacial Research.
We thank Pierre Rivailler and Fred Wang, Harvard University, for rhesus lymphocryptovirus DNA, Richard Longnecker, Northwestern University, for Daudi 29 cells, and Martin Muggeridge, Louisiana State University Health Sciences Center, for the pCAGGS/MCS and pST7-Luc vectors and for helpful discussion.
REFERENCES
Akula, S. M., N. P. Pramod, F. Z. Wang, and B. Chandran. 2002. Integrin 31 (CD49c/29) is a cellular receptor for Kaposi's sarcoma-associated herpesvirus (KSHV/HHV8) entry into target cells. Cell 108:407-419.
Anderson, R. A., and U. A. Gompels. 1999. N- and C-terminal external domains of human herpesvirus-6 glycoprotein H affect a fusion-associated conformation mediated by glycoprotein L binding the N terminus. J. Gen. Virol. 80:1485-1494.
Borza, C. M., and L. M. Hutt-Fletcher. 2002. Alternate replication in B cells and epithelial cells switches tropism of Epstein-Barr virus. Nat. Med. 8:594-599.
Borza, C. M., A. J. Morgan, S. M. Turk, and L. M. Hutt-Fletcher. 2004. Use of gHgL for attachment of Epstein-Barr virus to epithelial cells compromises infection. J. Virol. 78:5007-5014.
Cho, Y.-G., J. Ramer, P. Rivailler, C. Quink, R. L. Gerber, D. R. Beier, and F. Wang. 2001. An Epstein-Barr-related herpesvirus from marmoset lymphomas. Proc. Natl. Acad. Sci. USA 98:1224-1229.
Ferrer, M., T. M. Kapoor, T. Strassmaier, W. Weissenhorn, J. J. Skehel, D. Oprian, S. L. Schreiber, D. C. Wiley, and S. C. Harrison. 1999. Selection of gp41 mediated HIV-1 cell entry inhibitors from biased combinatorial libraries of non-natural binding elements. Nat. Struct. Biol. 6:953-960.
Galdiero, M., A. Whiteley, B. Bruun, S. Bell, T. Minson, and H. Browne. 1997. Site-directed and linker insertion mutagenesis of herpes simplex virus type 1 glycoprotein H. J. Virol. 71:2163-2170.
Gianni, T., P. L. Martelli, R. Casadio, and G. Campadelli-Fiume. 2005. The ectodomain of herpes simplex virus glycoprotein H contains a membrane alpha-helix with attributes of an internal fusion peptide, positionally conserved in the Herpesviridae family. J. Virol. 79:2931-2940.
Haan, K. M., S. K. Lee, and R. Longnecker. 2001. Different functional domains in the cytoplasmic tail of glycoprotein gB are involved in Epstein-Barr virus induced membrane fusion. Virology 290:106-114.
Haddad, R. S., and L. M. Hutt-Fletcher. 1989. Depletion of glycoprotein gp85 from virosomes made with Epstein-Barr virus proteins abolishes their ability to fuse with virus receptor-bearing cells. J. Virol. 63:4998-5005.
Heineman, T., M. Gong, J. Sample, and E. Kieff. 1988. Identification of the Epstein-Barr virus gp85 gene. J. Virol. 62:1101-1107.
Li, Q. X., M. K. Spriggs, S. Kovats, S. M. Turk, M. R. Comeau, B. Nepom, and L. M. Hutt-Fletcher. 1997. Epstein-Barr virus uses HLA class II as a cofactor for infection of B lymphocytes. J. Virol. 71:4657-4662.
Li, Q. X., S. M. Turk, and L. M. Hutt-Fletcher. 1995. The Epstein-Barr virus (EBV) BZLF2 gene product associates with the gH and gL homologs of EBV and carries an epitope critical to infection of B cells but not of epithelial cells. J. Virol. 69:3987-3994.
Liu, D. X., U. A. Gompels, L. Foa-Tomasi, and G. Campadelli-Fiumi. 1993. Human herpesvirus 6 glycoprotein H and L homologues are components of the gp100 complex and the gH external domain is the target for neutralizing monoclonal antibodies. Virology 197:12-22.
McShane, M. P., and R. Longnecker. 2004. Cell-surface expression of a mutated Epstein-Barr virus glycoprotein B allows fusion independent of other viral proteins. Proc. Natl. Acad. Sci. USA 101:17474-17479.
Miller, N., and L. M. Hutt-Fletcher. 1992. Epstein-Barr virus enters B cells and epithelial cells by different routes. J. Virol. 66:3409-3414.
Miller, N., and L. M. Hutt-Fletcher. 1988. A monoclonal antibody to glycoprotein gp85 inhibits fusion but not attachment of Epstein-Barr virus. J. Virol. 62:2366-2372.
Molesworth, S. J., C. M. Lake, C. M. Borza, S. M. Turk, and L. M. Hutt-Fletcher. 2000. Epstein-Barr virus gH is essential for penetration of B cell but also plays a role in attachment of virus to epithelial cells. J. Virol. 74:6324-6332.
Moss, B., O. Elroy-Stein, T. Mizukami, W. A. Alexander, and T. R. Fuerst. 1990. New mammalian expression vectors. Nature 348:91-92.
Niwa, H., K. Yamamura, and J. Miyazaki. 1991. Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108:193-199.
Pegtel, D. M., J. Middeldorp, and D. A. Thorley-Lawson. 2004. Epstein-Barr virus infection in ex vivo tonsil epithelial cell cultures of asymptomatic carriers. J. Virol. 78:12613-12624.
Rickinson, A. B., and E. Kieff. 2001. Epstein-Barr virus, p. 2575-2627. In D. M. Knipe and P. M. Howley (ed.), Fields virology, vol. 2. Lippincott Williams and Wilkins, Philadelphia, Pa.
Rivailler, P., H. Jiang, Y.-G. Cho, C. Quink, and F. Wang. 2002. Complete nucleotide sequence of the rhesus lymphocryptovirus: genetic validation for an Epstein-Barr virus animal model. J. Virol. 76:421-426.
Spear, P. G., and R. Longnecker. 2003. Herpesvirus entry: an update. J. Virol. 77:10179-10185.
Spriggs, M. K., R. J. Armitage, M. R. Comeau, L. Strockbine, T. Farrah, B. MacDuff, D. Ulrich, M. R. Alderson, J. Mullberg, and J. I. Cohen. 1996. The extracellular domain of the Epstein-Barr virus BZLF2 protein binds the HLA-DR beta chain and inhibits antigen presentation. J. Virol. 70:5557-5563.
Thorley-Lawson, D. A., and A. Gross. 2004. Persistence of Epstein-Barr virus and the origins of associated lymphomas. N. Engl. J. Med. 350:1328-1337.
Tugizov, S. M., J. W. Berline, and J. M. Palefsky. 2003. Epstein-Barr virus infection of polarized tongue and nasopharyngeal epithelial cells. Nat. Med. 9:307-314.
Wang, X., and L. M. Hutt-Fletcher. 1998. Epstein-Barr virus lacking glycoprotein gp42 can bind to B cells but is not able to infect. J. Virol. 72:158-163.
Wang, X., W. J. Kenyon, Q. X. Li, J. Mullberg, and L. M. Hutt-Fletcher. 1998. Epstein-Barr virus uses different complexes of glycoproteins gH and gL to infect B lymphocytes and epithelial cells. J. Virol. 72:5552-5558.
Whetter, L. E., S. P. Day, E. A. Brown, O. Elroy-Stein, and S. M. Lemon. 1994. Analysis of hepatitis A virus translation in a T7 polymerase-expressing cell line. Arch. Virol. Suppl. 9:291-298.
Yaswen, L. R., E. B. Stephens, L. C. Davenport, and L. M. Hutt-Fletcher. 1993. Epstein-Barr virus glycoprotein gp85 associates with the BKRF2 gene product and is incompletely processed as a recombinant protein. Virology 195:387-396.(Liguo Wu, Corina M. Borza)
Department of Microbiology and Immunology, Center for Molecular and Tumor Virology, and Feist-Weiller Cancer Center, Louisiana State University Health Sciences Center, Shreveport, Louisiana
ABSTRACT
The core fusion machinery of all herpesviruses consists of three conserved glycoproteins, gB and gHgL, suggesting a common mechanism for virus cell fusion, but fusion of Epstein-Barr virus (EBV) with B cells and epithelial cells is initiated differently. Fusion with B cells requires a fourth protein, gp42, which complexes with gHgL and interacts with HLA class II, the B-cell coreceptor. Fusion with an epithelial cell does not require gp42 but requires interaction of gHgL with a novel epithelial cell coreceptor. Epithelial cell fusion can be inhibited by gp42 binding to gHgL and by antibodies to gH that fail to block B-cell fusion. This suggests that regions of gHgL initiating fusion with each cell are separable from each other and from regions involved in fusion itself. To address this possibility we mapped the region of gH recognized by a monoclonal antibody to gH that blocks EBV fusion with epithelial cells but not B cells by making a series of chimeras with the gH homolog of rhesus lymphocryptovirus. Proteins with mutations engineered within this region included those that preferentially mediate fusion with B cells, those that preferentially mediate fusion with epithelial cells, and those that mediate fusion with neither cell type. These results support the hypothesis that the core fusion function of gH is the same for B cells and epithelial cells and that it differs only in the way in which it is triggered into a functionally active state.
INTRODUCTION
Epstein-Barr virus (EBV) is a human herpesvirus that predominantly infects B lymphocytes and epithelial cells. It causes infectious mononucleosis, an immunopathology initiated by an expansion of virus-infected B cells, immunoblastic lymphomas of the immunosuppressed, which are driven by virus infection, and oral hairy leukoplakia, an epithelial lesion produced as a result of lytic replication. The virus is also associated with additional lymphoid and epithelial malignancies, including Burkitt's lymphoma, Hodgkin's disease, and nasopharyngeal and gastric carcinomas (22). In normal healthy individuals the B cell is the primary reservoir of infection, but there is mounting evidence that epithelial cells are at least periodically infected during long-term carriage (21). Shuttling of virus between the two cell types may play a role in efficient initiation of infection of a new host and in replenishment of the infected B-cell pool (3, 26, 27).
Infection of B cells and epithelial cells occurs by different routes (16), and the subset of envelope proteins involved in mediating virus fusion with the two cell types is different. Following attachment via gp350 and CR2, fusion with a B cell requires glycoprotein gB and an oligomeric three-part complex of glycoproteins, gH, its chaperone gL, and gp42 (9). gB and gHgL are essential to the fusion machinery of all known herpesviruses (24); gp42 and its homologs are unique to EBV and other 1 herpesviruses (5, 13, 23). gp42 interacts with HLA class II on the B-cell surface (25), and this interaction is essential for triggering the fusion machinery into an active state (9, 10, 12, 13, 17, 28).
The requirements for fusion with an epithelial cell are less well understood but are in some respects fundamentally different. The core fusion machinery comprised of gHgL and gB is probably the same (15), but an interaction between gp42 and HLA class II is not used for its activation. Instead, gH, as part of a two-part gHgL oligomer, interacts directly with an as-yet-unknown molecule(s) in the epithelial membrane. This interaction, and fusion of virus with an epithelial cell, is impeded by the presence of gp42 or by a number of monoclonal antibodies (MAbs) to gH or gHgL that have no effect on infection of B cells (4, 18, 29). Thus, while only three-part complexes that include gp42 can participate in fusion with B cells, only two-part complexes that lack gp42 can participate in epithelial cell fusion.
EBV virions carry both three-part and two-part complexes, and modulation of the amounts of gp42 in the virion, which alters the ratio between the two, influences the tropism of the virus. Virions made in HLA class II-positive B cells lose some three-part complexes as a result of an intracellular interaction with HLA class II that targets the complexes to a degradative pathway. Virions made in HLA class II-negative epithelial cells lose none and carry a higher amount of gp42. Epithelial-derived virions are considerably better at infecting B cells, whereas B-cell-derived virions are somewhat better at infecting epithelial cells (3, 4).
If the core fusion machinery of EBV for both cell types were comprised of gB and gHgL and differed only in the way in which gHgL is triggered into a functionally active form, then it should be possible to derive a gHgL complex that is capable of supporting fusion with an epithelial cell but not a B cell, or vice versa. To test this hypothesis we first mapped the region of gH that is recognized by one of the MAbs that blocks epithelial but not B-cell infection by taking advantage of its failure to react with the gH homolog of the closely related rhesus lymphocryptovirus (Rh-LCV). A series of insertion mutations was made in this region in EBV gH, and each was tested for expression and the ability to mediate fusion with a B-cell or an epithelial cell membrane. We report here on the phenotypes of these mutated proteins, which include those that, although expressed at the cell surface, have either completely or only differentially lost the ability to mediate fusion with B cells and epithelial cells.
MATERIALS AND METHODS
Cells. CV-1 monkey kidney cells were grown in Dulbecco's modified Eagle's medium. AGS, a human gastric carcinoma cell line, and Chinese hamster ovary cells (CHO-K1) were grown in Ham's F-12 medium. Daudi 29 cells (a gift of Richard Longnecker, Northwestern University), B cells that stably express T7 RNA polymerase from the pOS2 vector (30), were grown in RPMI medium. All culture media were obtained from Gibco-BRL Life Technologies (Grand Island, NY) and supplemented with 10% heat-inactivated fetal bovine serum.
Antibodies. Antibodies used were MAbs E1D1 to gHgL, CL59 and CL40 to gH, CL55 to gB, F-2-1 to gp42 (13), and M2 (Sigma-Aldrich, St. Louis, MO) to the Flag epitope. All noncommercial antibodies were affinity purified on protein A-agarose.
Virus. Stocks of recombinant vaccinia virus expressing the T7 RNA polymerase (vvT7; a gift of William Britt, University of Alabama, Birmingham) were grown in CV-1 cells infected at a multiplicity of infection of 0.01. When all cells appeared swollen (approximately 48 h postinfection), they were shaken or scraped into the medium, pelleted by centrifugation, and resuspended in 2 ml of clarified medium per 30 million cells. The resuspended cells were subjected to three cycles of freeze-thawing, sonicated for 45 s in 4-ml aliquots in round-bottom tubes, clarified by centrifugation, pooled, and stored at –85°C.
Plasmid constructs for epitope mapping. Plasmids pTM1-EBV gH and pTM1-EBV gL (13) were made by cloning PCR-amplified sequences into the pTM1 vector (19), which contains a T7 promoter, the encephalomyocarditis virus cap-independent translation signal, a multiple cloning site, and a T7 transcriptional terminator. Translation initiates at the NcoI site in the multiple cloning region. In pTM1-EBV gH this site is lost as a result of blunt-end cloning. Plasmid pTM1-EBV119-Rh-LCV gH was made by cutting a 2.8-kb HindIII fragment encompassing the open reading frame encoding the gH homolog of Rh-LCV from the DK12 cosmid clone of Rh-LCV DNA (a gift of Fred Wang, Harvard University) (23). The fragment was blunt ended, recut with NcoI, and cloned into pTM1-EBV gH cut with NcoI and HincII. The first 119 bp of the insert represented EBV gH sequence, and the remaining 2,004 bp represented Rh-LCV gH sequence, including the stop codon and an additional 23 bp of 3' sequence. The EBV gH amino acid sequence encoded by the 119 bp of EBV DNA differed from Rh-LCV gH at positions 2, 7, 10, 13, 15, 17, 33, and 35. Only residues 33 and 35 are predicted to remain after cleavage of the signal sequence at residue 17. These residues were changed to those of Rh-LCV gH by amplification of a fragment extending from the NcoI site to a ClaI site upstream of the gH start in pTM1-EBV gH. The changes were incorporated in the primer encompassing the NcoI site to make pTM1-Rh-LCV gH. Further chimeras of EBV gH and Rh-LCV gH were constructed by insertion of PCR-amplified fragments of EBV DNA in place of equivalent fragments of Rh-LCV gH either by using existing shared restriction sites (NcoI at bp 119) or by creating a new site in the EBV gH DNA sequence, which did not alter coding sequence, to correspond with existing sites of Rh-LCV gH.
Transfection-infection protocol. CV-1 cells were grown to 90% confluence in 100-mm-diameter petri dishes and infected with vvT7 at a multiplicity of infection of 5. Thirty minutes later, the inoculum was removed, and the cells were washed twice in medium without serum and transfected with pTM1 plasmids. Ten micrograms of DNA was mixed with 30 μl of Lipofectamine (Gibco-BRL) made to a total volume of 200 μl with serum-free medium.
EBV plasmid constructs for fusion assays. EBV gH, EBV gL, EBV gB, and EBV gp42 were amplified by PCR and cloned into the pCAGGS/MCS vector (a gift of Martin Muggeridge, Louisiana State University Health Sciences Center) for expression under control of the -actin promoter in cooperation with the human cytomegalovirus immediate-early enhancer (20). These constructs were designated pCAGGS-gH, pCAGGS-gL, pCAGGS-gB, and pCAGGS-gp42, respectively. A Flag-tagged version of pCAGGS-gH, pCAGGS-Flag-gH, was made by insertion of an eight-amino-acid Flag epitope between residues 22 and 23 of the EBV gH sequence, four residues after the predicted cleavage site of the putative signal sequence. For construction of mutations in pCAGGS-Flag-gH, a 1.6-kb BglII-PvuII fragment encompassing sequence coding for residues 500 to 629 was cloned into the 2.5-kb pSP72 vector (Promega, Madison, WI) to make pSP72-EBV gH1.6. This construct was used as the template for a series of insertion mutants made using the QuikChange II protocol (Stratagene, La Jolla, CA). Primers for insertion mutagenesis were designed to introduce in-frame insertions of the sequence TCG CGA, which encodes serine and arginine residues and introduces a novel and unique NruI restriction site. DNA was amplified for 18 cycles (60 s at 95°C, 90 s at 60°C, and 9 min at 72°C), and the template was eliminated by digestion with DpnI. Insertions were initially screened by digestion with NruI. XhoI-NsiI fragments encompassing the mutated BglII-PvuII fragments were recloned back into pCAGGS-Flag-gH for assay of fusion activity or pTM1-EBV gH for biochemical analysis. The desired insertion or mutation was confirmed by sequencing of the fragments returned to pCAGGS-Flag-gH.
Immunofluorescence. For internal staining, slides bearing air-dried acetone-fixed cells were incubated at 37°C for 30 min in a humidified atmosphere with MAbs, washed three times with phosphate-buffered saline for 5 min each, reincubated for 30 min with the appropriate dilution of fluorescein isothiocyanate-conjugated sheep anti-mouse immunoglobulin (ICN/Cappel, Irvine, CA), washed three times, and mounted in mounting medium (Kirkegaard & Perry Laboratories, Gaithersburg, MD). For surface staining, unfixed cells were sequentially incubated with MAbs and fluorescein isothiocyanate-conjugated anti-mouse immunoglobulin. Cells were washed three times between incubations and three times before mounting in mounting medium.
Levels of cell surface expression. The levels of expression of wild-type and mutant gH constructs at the cell surface were measured by flow cytometric analysis of CHO-K1 cells transfected with pCAGGS-Flag-gH or gH mutants and pCAGGS-gL. A 3.2-μg aliquot of DNA was mixed with 4 μl Target transfectin F2 and 12 μl Target peptide enhancer (Targeting Systems, Santee, CA) in high-glucose Dulbecco's modified Eagle's medium. Twenty-four hours later, transfected cells were washed with ice-cold phosphate-buffered saline containing 2% fetal bovine serum and serially reacted with MAb M2 (anti-Flag) and phycoerythrin-conjugated anti-mouse immunoglobulin (Jackson ImmunoResearch Laboratories, West Grove, PA), with washing between each step.
AGS cell-cell fusion assay. AGS cells were seeded in two-well chamber slides and were transfected at 70 to 80% confluence for 4 h with 0.25 μg pCAGGS-Flag-gH or 0.25 μg mutated pCAGGS-Flag-gH together with 0.25 μg pCAGGS-gL and 0.6 μg pCAGGS-gB. Plasmids were mixed with 1 μl Target transfectin F2 and 3 μl Target peptide enhancer in high-glucose Dulbecco's modified Eagle's medium. Twenty-four hours posttransfection cells were fixed with ice-cold acetone and stained with MAb to gB. Cells expressing gB and containing four or more nuclei were considered to have undergone fusion and were recorded as one fusion event. The extent of fusion was calculated as the number of fusion events as a percentage of the total number cells of expressing gB. No fusion was seen in cells that did not stain with MAb to gB.
Epithelial cell-B-cell fusion assay. CHO-K1 cells were seeded in six-well plates and were transfected at 70 to 80% confluence for 4 h with 0.8 μg pCAGGS-Flag-gH or mutated pCAGGS-Flag-gH together with 0.8 μg gL, 1.4 μg pCAGGS-gB, and 1.2 μg pCAGGSgp42 as well as 0.8 μg pST7-Luc containing the T7 promoter upstream of the luciferase gene (6) (a gift of Martin Muggeridge). Each well of transfected cells was then overlaid with two million Daudi 29 cells that expressed T7 RNA polymerase. Twenty-four hours later, cells were washed twice with phosphate-buffered saline and lysed with 500 μl passive lysis buffer (Promega). Luciferase substrate (100 μl) was added to 20 μl supernatant of lysate. Luminosity readings were obtained by using a TD-20/20 luminometer (Promega).
Radiolabeling and immunoprecipitation. CV-1 cells infected with vvT7 and transfected with pTM1 plasmids were labeled biosynthetically with 100 μCi of Pro-Mix (70% [35S]methionine, 30% [35S]cysteine) and 300 μCi of [35S]cysteine (>1,000 Ci/mmol; Amersham Biosciences, Arlington, IL) per dish (approximately 107 cells) as previously described (13). Labeled cells were solubilized in radioimmunoprecipitation buffer (50 mM Tris-HCl, pH 7.2, 0.15 M NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate [SDS], 0.1 mM phenylmethylsulfonyl fluoride, and 100 U of aprotinin per ml) and immunoprecipitated with antibody and protein A-Sepharose CL4B (Sigma). Immu-noprecipitated proteins were washed, dissociated by boiling for 2 min in sample buffer with 2-mercaptoethanol, and analyzed by SDS-polyacrylamide gel electrophoresis in 10% polyacrylamide cross-linked with 0.28% N,N'-diallyltartardiamide followed by fluorography.
RESULTS
The region of gH that interacts with MAb CL59 lies between residues 500 and 629. MAbs E1D1, CL40, and CL59 all inhibit EBV infection of epithelial cells and, since none of the antibodies neutralize B-cell infection, the epitopes that they recognize are uniquely important to infection of epithelial cells (13, 18). MAb E1D1 reacts with EBV gH only if it is expressed together with EBV gL, and MAbs CL40 and CL59 react with gH alone. To begin to map the regions of gH involved in the generation of these epitopes, we explored the reactivity of the MAbs with the gH homolog of Rh-LCV, which has 86% similarity to EBV gH (23). EBV gL was expressed together with Rh-LCV gH in the pTM1 vector for expression under control of the T7 promoter. The plasmids were transfected into CV1 cells that were also infected with recombinant vaccinia virus expressing the T7 polymerase. Transfected and infected cells were acetone fixed and tested for reactivity with MAbs in indirect immunofluorescence assays. MAbs E1D1 and CL40 both reacted with Rh-LCV-gH, but MAb CL59 did not (Fig. 1).
In order to map the domain recognized by MAb CL59, a series of chimeric constructs were made in which different fragments of EBV gH replaced linearly equivalent fragments of Rh-LCV gH. The ability of each chimera to be recognized by the antibody was examined by indirect immunofluorescence. EBV residues 1 to 39, 40 to 319, or 320 to 457 did not confer reactivity with MAb CL59, but residues 458 to 628 did (Fig. 1), indicating that the MAb CL59 epitope resides within this fragment of EBV gH (Fig. 2). Reactivity was retained when only EBV gH residues 501 to 628, which have 74.7% identity with the equivalent Rh-LCV gH fragment (Fig. 3), were replaced. In an attempt to further narrow down the region recognized by CL59, two final chimeric constructs were made, one in which only residues 501 to 589 were replaced with EBV sequence and one in which only residues 590 to 628 were replaced. Neither of these two constructs was recognized by MAb CL59. It is formally possible that the epitope recognized by MAb CL59 is linear and spans the junction of the two fragments. However, since MAb fails to react with EBV gH on Western blots (data not shown), it is perhaps more likely that it recognizes a nonlinear epitope formed by folding of residues 501 to 628.
Expression of EBV gH with insertion mutations between residues 500 and 629. Mapping of the reactivity of MAb CL59 to residues 501 to 628 of EBV gH identified a region that was potentially of unique importance to epithelial infection. To test this possibility a series of 13 insertion mutations were made within the region (Fig. 3) and built into an EBV gH construct that had been cloned into the pCAGGS vector for expression under control of the -actin promoter (pCAGGS-Flag-gH). Because it was possible that some of the insertions might affect the ability of the protein to be recognized by all of the gH-specific MAbs, sequence encoding a linear Flag epitope was also inserted at a position that would put it two residues after the predicted cut site of the putative signal sequence (11). Each mutational insertion consisted of two amino acids, serine and arginine, since a similar strategy has proven viable for identifying functional regions of another conformationally sensitive herpesvirus glycoprotein, glycoprotein gB of herpes simplex virus type 2 (M. Muggeridge, personal communication).
The expression of each construct, together with pCAGGS-gL, was first examined by indirect immunofluorescence staining of acetone-fixed cells. Although all constructs were expressed as judged by reactivity with the anti-Flag antibody, they fell phenotypically into three groups (Table 1). Numbers 1 to 6 and 9 to 11 reacted well not only with anti-Flag but also with all three EBV-specific MAbs. Numbers 7 and 13 reacted with the same antibodies but more weakly. Numbers 8 and 12, which also reacted only weakly with anti-Flag antibody, failed to react at all with any of the EBV-specific MAbs. To determine which, if any, of the mutants could be expressed at the cell surface, indirect immunofluorescence staining was repeated with unfixed cells. The constructs again fell into three broad groups (Table 2). Group one mutants, including numbers 1 to 5 and 9 to 11, were readily detected at wild-type levels at the cell surface with both anti-Flag and all the EBV-specific MAbs. Group two mutants, including number 6, which had reacted well in fixed cells, and numbers 7 and 13, all reacted only weakly at the cell surface with anti-Flag and the EBV-specific MAbs. Number 6 contained an insertion that interrupted an N-linked glycosylation site and had a faster mobility when analyzed by electrophoresis and Western blotting (not shown). Group three, including numbers 8 and 12, was not seen by anti-Flag or any of the EBV-specific MAbs.
A more quantitative assessment of the relative levels of expression at the cell surface of each of the constructs made by flow cytometric analysis of unfixed cells stained with M2 anti-Flag antibody revealed some more subtle heterogeneities. Group one could now be divided into numbers 3, 5, 9, and 10, which were expressed at levels similar to wild-type pCAGGS-Flag-gH, and numbers 1, 2, 4, and 11, which were expressed less well but still at more than 50% of wild-type levels (Fig. 4a). Group two, numbers 6, 7, and 13, was expressed at approximately 25% of wild-type levels. As before, no cell surface expression was seen of insertion mutants numbers 8 and 12, both of which failed to react with any of the EBV-specific MAbs.
Comparison of the ability of wild-type gH and insertion mutants to mediate cell-to-cell fusion of epithelial cells. pCAGGS-Flag-gH and each of the insertion mutants were next examined for the ability to mediate cell-to-cell fusion when transfected into AGS epithelial cells together with pCAGGS-gL and pCAGGS-gB. As expected, no fusion was seen in cells transfected with pCAGGS-gB alone, cells transfected with pCAGGS-Flag-gH and pCAGGS-gL, or cells transfected with pCAGGS-gB and pCAGGS-gL. However, when pCAGGS-gB, pCAGGS-Flag-gH, and pCAGGS-gL were coexpressed, multinucleate cells were seen (Fig. 5). A count of the number of cells that were successfully transfected as indicated by reactivity with MAb CL55 to gB and the number of foci of fluorescence in which more than four nuclei were present indicated that pCAGGS-Flag-gH supported fusion of 49 ± 2% (mean ± standard deviation) of the total number of transfected cells. A pCAGGS-gH construct expressing wild-type gH without insertion of the Flag tag mediated a similar degree of fusion (data not shown).
The fusion mediated by each insertion mutant was expressed as a percentage of fusion supported by wild-type pCAGGS-Flag-gH. None induced fusion as efficiently as wild-type gH, but numbers 3, 5, and 10, all of which were expressed at wild-type levels, induced fusion to a level of at least 50% of wild-type pCAGGS-Flag-gH (Fig. 4b). Numbers 1 and 2, which were expressed at more than 50% of wild-type levels, and number 13, which was expressed at about 25% of wild-type levels, induced lower but still significant levels of fusion. Numbers 4, 6, 7, 8, 9, 11, and 12 induced little or no fusion despite the fact that all but two, 8 and 12, were recognized by MAbs E1D1, CL40, and CL59 and two, numbers 9 and 11, were expressed at wild-type or close-to-wild-type levels, respectively.
Comparison of the ability of wild-type gH and insertion mutants to mediate cell-cell fusion with target B cells. Fusion of B cells which grow in suspension is not readily detected by microscopy. As an indirect measure of the ability of wild-type gH and insertion mutants to mediate fusion with a B cell, we therefore used an assay previously developed for this purpose by Haan and colleagues (9). Constructs were transfected into CHO-K1 cells, which do not fuse after expression of EBV gHgL and gB (15), together with pCAGGS-gL, pCAGGS-gB, pCAGGS-gp42, and pST7-Luc. Transfected cells were subsequently overlaid with Daudi 29 B cells which expressed the T7 polymerase. Fusion was determined as the increase in luciferase activity relative to a control in which pCAGGS-Flag-gH or its mutated derivatives were replaced with empty vector. The degree of fusion mediated by each insertion mutant was then expressed as a percentage of that induced by wild-type pCAGGS-Flag-gH.
Three mutants, numbers 3, expressed at wild-type levels, number 11, expressed at approximately 75% of wild-type levels, and number 13, expressed at approximately 25% of wild-type levels, induced levels of fusion that were variable but in a series of 12 experiments proved to be not significantly different from wild type (Fig. 4c). Number 5, which was expressed at wild-type levels, supported fusion to a level of only 50% of wild type. Lower levels of fusion again were mediated by numbers 1 and 2, which were expressed at between 50 and 75% of wild-type levels. Numbers 6 to 10 and 12 supported little or no fusion. Of these, numbers 9 and 10 were expressed at wild-type levels, 6 and 7 were expressed at about 25% of wild type, and 8 and 12 were not detectable at the cell surface at all.
Interaction of insertion mutants with gp42. A comparison of the levels of cell surface expression and abilities to mediate fusion with epithelial cells or B cells highlighted three particularly interesting insertion mutants, numbers 9, 10, and 11, which, although expressed at wild-type or close-to-wild-type levels, had very different phenotypes. Number 10 preferentially supported fusion with epithelial cells, number 11 preferentially supported fusion with B cells, and number 9 supported fusion with neither. Wild-type gH, if associated with gL, forms a three-part complex of gH, gL, and gp42 and can be immunoprecipitated by MAb F-2-1 to gp42. To determine if the low levels of B-cell fusion mediated by numbers 9 and 10 reflected an inability to interact with gp42, the same mutations were built into pTM1-EBV gH for higher levels of expression. Cells transfected with wild-type or mutated pTM1-EBV gH together with pTM1 gL and pTM1 gp42 were radiolabeled, and proteins were immunoprecipitated with MAb F-2-1 to gp42. F-2-1 immunoprecipitated not only wild-type pTM1-EBV gH but also the mutated proteins, indicating that despite their compromised ability to support B-cell fusion they both retained the ability to reform the three-part complex of gH, gL, and gp42 (Fig. 6).
DISCUSSION
Entry of all herpesviruses into cells requires fusion of the virus envelope with a cell membrane, either at the cell surface or after endocytosis. The ways in which this is accomplished are not completely understood for any herpesvirus, but some common themes are emerging (reviewed in reference 24). Three glycoproteins, gB, gH, and gL, which have homologs in every herpesvirus so far studied, are essential and make up the core fusion machinery. Triggering this machinery into action requires interaction with a coreceptor or entry mediator. Thus, triggering fusion of herpes simplex virus with many cells requires an interaction between a fourth glycoprotein, glycoprotein gD, and one of several cell entry receptors. Triggering of fusion of human herpesvirus 8 probably requires an interaction between gB and integrins (1). EBV is, however, perhaps unusual in two ways. First, although fusion with a B cell requires a fourth protein, gp42, which interacts with the entry mediator or coreceptor HLA class II, fusion with an epithelial cell does not. Second, gp42, which is essential for B-cell fusion, interferes with epithelial cell fusion, and MAbs to gHgL that block epithelial cell fusion have no effect on fusion with a B cell (4, 29). The understanding that gHgL are essential for fusion with both B cells and epithelial cells, despite the fact that the activation of fusion is different, was our impetus to look for regions of the complex that are important to function as part of the core fusion machinery and regions that are separably involved in triggering B cell and epithelial cell fusion. Since we had made three MAbs that inhibited epithelial cell entry but not B-cell entry, mapping their reactivities seemed a rational approach to begin the search for such regions. We were fortunate to find not only that only two of the three MAbs reacted with the gH homolog of the very closely related Rh-LCV when it was expressed together with EBV gL, but also that the MAb that failed to react was one that recognizes EBV gH in the absence of gL. Its epitope could then conclusively be said to be carried by gH and chimeras of Rh-LCV gH, and EBV gH could be used locate its general position. The region of gH carrying the epitope, which we were unable to narrow down further than a 129-amino-acid stretch, was in the carboxy-terminal third of the protein, only 52 amino acids from the predicted transmembrane domain. It does not overlap with the equivalent domain of herpes simplex virus gH that has been identified as a potential fusion protein (8). However, it is a region of gH that has been implicated in fusion of both the alphaherpesvirus herpes simplex virus (7), and the betaherpesvirus human herpesvirus 6 (2, 14). The insertion mutations described here confirm its functional importance.
Three insertion mutants, numbers 9, 10, and 11, were of particular interest, because although each was expressed at wild-type or close-to-wild-types levels, each had a very different functional phenotype. Number 9 supported fusion with neither B cells nor epithelial cells and would appear to confirm that, as for other herpesviruses, this carboxy-terminal region of gH is critical for function of the core fusion machinery. Number 10 preferentially supported epithelial cell fusion, whereas number 11 supported wild-type levels of fusion with B cells and little or no fusion with epithelial cells. The failure of number 11 to support epithelial cell fusion could not simply be ascribed to its expression level, since other insertion mutants, such as numbers 1, 2, and 13, were expressed at similar or lower levels and supported significant fusion. This suggests that, as expected, the triggers provided by gp42 interacting with HLA class II and by gHgL interacting with the epithelial cell coreceptor are mediated by separate structural features of gHgL. Further analysis of mutant number 11 will be needed to determine if the insertion has interfered with the ability of virus to bind to epithelial cells via a gHgL coreceptor (4, 18). Since both insertion mutants numbers 9 and 10 retained their interaction with gp42, both are likely to have retained the ability to interact indirectly with the B-cell coreceptor HLA class II. The fairly close proximity of insertions 10 and 11 suggests that the structures required for a functional interaction with either a B cell or an epithelial cell coreceptor may be proximal. This would not be altogether surprising, since complexes of gHgL that are saturated with gp42 cannot support epithelial cell fusion (29) and a soluble form of gp42 that can bind in trans impedes the interaction between gHgL and a specific epithelial receptor, possibly the coreceptor needed for fusion (4). MAb CL59 fails to block B-cell fusion, and addition of soluble gp42 to virus does not reduce the ability of the virus to bind CL59, as judged by flow cytometry (unpublished data). However, this could be explained if the antibody had a lower affinity for gHgL than did gp42 and was thus unable to compete with gp42 for binding to the same region of the gHgL complex.
The majority of the insertion mutations that were made preserved the ability of the protein to be recognized by each of the MAbs specific for gHgL which recognize conformational epitopes. In addition, all of those recognized by these MAbs were expressed at least at some level at the cell surface. This suggests that each was able to interact with gL. The association was not formally evaluated, since it is well established for all herpesviruses, including EBV (31), that without such an interaction gH is not authentically processed and transported to the cell surface.
Not surprisingly, mutants that were not expressed at all at the cell surface were also unable to support fusion with epithelial cells, but beyond this there was no simple relationship between levels of expression and levels of fusion with epithelial cells. Thus, of the four mutants, numbers 3, 5, 9, and 10, which were expressed at wild-type levels, three supported fusion at greater than 50% of wild-type levels and one supported no fusion at all. None of the mutants, even those expressed at wild-type levels, were able to support epithelial cell fusion as extensively as the wild-type protein. It is impossible to know, without a structural model of gHgL, exactly what this reflects, but it suggests that perhaps even though the insertion mutants are authentically transported and are reactive with MAbs known to recognize conformational epitopes, they are at least to some degree either initially misfolded or unable, subsequently, to assume fully a conformation required for switching to a functionally active state.
In contrast, three of the insertion mutants, numbers 3, 11, and 13, supported fusion with B cells at levels that were not significantly different from wild type and another, number 5, supported fusion at a level greater than 50% of wild type. The assays used to estimate fusion with the two different cell types are clearly different. It is formally possible that the luciferase assay is in some cases measuring pore formation in the absence of the pore expansion that would be necessary for production of multinucleate cells that were scored as epithelial fusion. However, the luciferase assay is not simply overestimating the ability of a given protein to support fusion since, as is clear from the composite analysis, there were two insertion mutants, numbers 5 and 10, which scored higher in the epithelial assay than the B-cell assay.
In the absence of a crystal structure for gHgL, it is not possible to understand the full impact of the mutations that have been made. However, three points seem clear. First, that the carboxy-terminal region of gH is critical to the support of fusion with either B cells or epithelial cells and is thus relevant to the function of the core fusion machinery. Second, that the same region is relevant to the trigger by which gHgL becomes functional. Third, that although mutations in this region can interfere with triggering via HLA class II or the epithelial cell coreceptor, the sequences important for the triggering events are distinct.
ACKNOWLEDGMENTS
This work was supported by Public Health Services grants AI20662 from the National Institute of Allergy and Infectious Diseases and DE016669 from the National Institute of Dental and Craniofacial Research.
We thank Pierre Rivailler and Fred Wang, Harvard University, for rhesus lymphocryptovirus DNA, Richard Longnecker, Northwestern University, for Daudi 29 cells, and Martin Muggeridge, Louisiana State University Health Sciences Center, for the pCAGGS/MCS and pST7-Luc vectors and for helpful discussion.
REFERENCES
Akula, S. M., N. P. Pramod, F. Z. Wang, and B. Chandran. 2002. Integrin 31 (CD49c/29) is a cellular receptor for Kaposi's sarcoma-associated herpesvirus (KSHV/HHV8) entry into target cells. Cell 108:407-419.
Anderson, R. A., and U. A. Gompels. 1999. N- and C-terminal external domains of human herpesvirus-6 glycoprotein H affect a fusion-associated conformation mediated by glycoprotein L binding the N terminus. J. Gen. Virol. 80:1485-1494.
Borza, C. M., and L. M. Hutt-Fletcher. 2002. Alternate replication in B cells and epithelial cells switches tropism of Epstein-Barr virus. Nat. Med. 8:594-599.
Borza, C. M., A. J. Morgan, S. M. Turk, and L. M. Hutt-Fletcher. 2004. Use of gHgL for attachment of Epstein-Barr virus to epithelial cells compromises infection. J. Virol. 78:5007-5014.
Cho, Y.-G., J. Ramer, P. Rivailler, C. Quink, R. L. Gerber, D. R. Beier, and F. Wang. 2001. An Epstein-Barr-related herpesvirus from marmoset lymphomas. Proc. Natl. Acad. Sci. USA 98:1224-1229.
Ferrer, M., T. M. Kapoor, T. Strassmaier, W. Weissenhorn, J. J. Skehel, D. Oprian, S. L. Schreiber, D. C. Wiley, and S. C. Harrison. 1999. Selection of gp41 mediated HIV-1 cell entry inhibitors from biased combinatorial libraries of non-natural binding elements. Nat. Struct. Biol. 6:953-960.
Galdiero, M., A. Whiteley, B. Bruun, S. Bell, T. Minson, and H. Browne. 1997. Site-directed and linker insertion mutagenesis of herpes simplex virus type 1 glycoprotein H. J. Virol. 71:2163-2170.
Gianni, T., P. L. Martelli, R. Casadio, and G. Campadelli-Fiume. 2005. The ectodomain of herpes simplex virus glycoprotein H contains a membrane alpha-helix with attributes of an internal fusion peptide, positionally conserved in the Herpesviridae family. J. Virol. 79:2931-2940.
Haan, K. M., S. K. Lee, and R. Longnecker. 2001. Different functional domains in the cytoplasmic tail of glycoprotein gB are involved in Epstein-Barr virus induced membrane fusion. Virology 290:106-114.
Haddad, R. S., and L. M. Hutt-Fletcher. 1989. Depletion of glycoprotein gp85 from virosomes made with Epstein-Barr virus proteins abolishes their ability to fuse with virus receptor-bearing cells. J. Virol. 63:4998-5005.
Heineman, T., M. Gong, J. Sample, and E. Kieff. 1988. Identification of the Epstein-Barr virus gp85 gene. J. Virol. 62:1101-1107.
Li, Q. X., M. K. Spriggs, S. Kovats, S. M. Turk, M. R. Comeau, B. Nepom, and L. M. Hutt-Fletcher. 1997. Epstein-Barr virus uses HLA class II as a cofactor for infection of B lymphocytes. J. Virol. 71:4657-4662.
Li, Q. X., S. M. Turk, and L. M. Hutt-Fletcher. 1995. The Epstein-Barr virus (EBV) BZLF2 gene product associates with the gH and gL homologs of EBV and carries an epitope critical to infection of B cells but not of epithelial cells. J. Virol. 69:3987-3994.
Liu, D. X., U. A. Gompels, L. Foa-Tomasi, and G. Campadelli-Fiumi. 1993. Human herpesvirus 6 glycoprotein H and L homologues are components of the gp100 complex and the gH external domain is the target for neutralizing monoclonal antibodies. Virology 197:12-22.
McShane, M. P., and R. Longnecker. 2004. Cell-surface expression of a mutated Epstein-Barr virus glycoprotein B allows fusion independent of other viral proteins. Proc. Natl. Acad. Sci. USA 101:17474-17479.
Miller, N., and L. M. Hutt-Fletcher. 1992. Epstein-Barr virus enters B cells and epithelial cells by different routes. J. Virol. 66:3409-3414.
Miller, N., and L. M. Hutt-Fletcher. 1988. A monoclonal antibody to glycoprotein gp85 inhibits fusion but not attachment of Epstein-Barr virus. J. Virol. 62:2366-2372.
Molesworth, S. J., C. M. Lake, C. M. Borza, S. M. Turk, and L. M. Hutt-Fletcher. 2000. Epstein-Barr virus gH is essential for penetration of B cell but also plays a role in attachment of virus to epithelial cells. J. Virol. 74:6324-6332.
Moss, B., O. Elroy-Stein, T. Mizukami, W. A. Alexander, and T. R. Fuerst. 1990. New mammalian expression vectors. Nature 348:91-92.
Niwa, H., K. Yamamura, and J. Miyazaki. 1991. Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108:193-199.
Pegtel, D. M., J. Middeldorp, and D. A. Thorley-Lawson. 2004. Epstein-Barr virus infection in ex vivo tonsil epithelial cell cultures of asymptomatic carriers. J. Virol. 78:12613-12624.
Rickinson, A. B., and E. Kieff. 2001. Epstein-Barr virus, p. 2575-2627. In D. M. Knipe and P. M. Howley (ed.), Fields virology, vol. 2. Lippincott Williams and Wilkins, Philadelphia, Pa.
Rivailler, P., H. Jiang, Y.-G. Cho, C. Quink, and F. Wang. 2002. Complete nucleotide sequence of the rhesus lymphocryptovirus: genetic validation for an Epstein-Barr virus animal model. J. Virol. 76:421-426.
Spear, P. G., and R. Longnecker. 2003. Herpesvirus entry: an update. J. Virol. 77:10179-10185.
Spriggs, M. K., R. J. Armitage, M. R. Comeau, L. Strockbine, T. Farrah, B. MacDuff, D. Ulrich, M. R. Alderson, J. Mullberg, and J. I. Cohen. 1996. The extracellular domain of the Epstein-Barr virus BZLF2 protein binds the HLA-DR beta chain and inhibits antigen presentation. J. Virol. 70:5557-5563.
Thorley-Lawson, D. A., and A. Gross. 2004. Persistence of Epstein-Barr virus and the origins of associated lymphomas. N. Engl. J. Med. 350:1328-1337.
Tugizov, S. M., J. W. Berline, and J. M. Palefsky. 2003. Epstein-Barr virus infection of polarized tongue and nasopharyngeal epithelial cells. Nat. Med. 9:307-314.
Wang, X., and L. M. Hutt-Fletcher. 1998. Epstein-Barr virus lacking glycoprotein gp42 can bind to B cells but is not able to infect. J. Virol. 72:158-163.
Wang, X., W. J. Kenyon, Q. X. Li, J. Mullberg, and L. M. Hutt-Fletcher. 1998. Epstein-Barr virus uses different complexes of glycoproteins gH and gL to infect B lymphocytes and epithelial cells. J. Virol. 72:5552-5558.
Whetter, L. E., S. P. Day, E. A. Brown, O. Elroy-Stein, and S. M. Lemon. 1994. Analysis of hepatitis A virus translation in a T7 polymerase-expressing cell line. Arch. Virol. Suppl. 9:291-298.
Yaswen, L. R., E. B. Stephens, L. C. Davenport, and L. M. Hutt-Fletcher. 1993. Epstein-Barr virus glycoprotein gp85 associates with the BKRF2 gene product and is incompletely processed as a recombinant protein. Virology 195:387-396.(Liguo Wu, Corina M. Borza)